Book Review — How to Make the World Add Up: Ten Rules for Thinking Differently About Numbers by Tim Harford

A file photograph of share buying and selling at a agency in Kolkata (Specific photograph)

The months because the pandemic started have been a puzzle when decoding varied information on Covid, be it assessments, infections, deaths, vaccine efficacy and unintended effects and so forth. Media studies spin a brand new story on a regular basis, which frequently evokes the concern factor. That’s the downside with information and interpretation. Temper swings may be pushed by how we see such information that typically works on the negativity intuition. Due to this fact, you could “step again and benefit from the view”. This is among the 10 guidelines put ahead by Tim Harford in one other entertaining ebook on how to take a look at information or the warning to be exercised when decoding it.

Presentation of information is not only tough but additionally harmful in some nations, as Harford exhibits us. In Tanzania criticising official information is a prison offence. Graciela Bevacqua of Argentina was requested to indicate low inflation in 2007 and non-compliance meant getting the sack. In 2010 Andreas Georgiou of Greece was pressured to indicate a decrease fiscal deficit and when he got here up with plenty of 15.4%, he was faraway from the statistics division. One other precept of Harford right here is that “don’t take statistical bedrock without any consideration”. We in India will really feel nearer to this rule as all the info on employment, revision in methodology of calculating GDP and so forth has ignited comparable debate.

The issue with statistics is that it might probably show something. Early on it was discovered that nations that had excessive inhabitants additionally had massive variety of storks and therefore the 2 had been linked with causality. Then this was dismissed as being nonsense. The identical adverse argument was used to say that merely as a result of numerous people who smoke had been detected with most cancers, it didn’t show something. The tobacco foyer now had a robust argument on their aspect!

It’s with these tales and examples that Harford takes us by the pitfalls of drawing conclusions based mostly on information. At instances we really feel emotionally sturdy towards one thing and are affected by the ‘ostrich impact’. Can one consider the most typical instance? Sure, it’s the inventory market. When markets increase, we refuse to imagine that there’s something amiss in Covid instances and discover some rationalisation for the brand new highs, as consultants inform us issues will get even higher. It’s therefore fairly bizarre that a complete market may be interpreted on this method as we’re seeing right now, to the extent that everybody believes these actions and it turns into self-fulfilling.

Typically, we’ve surveys or experiments which can be carried out after which generalised, and this holds for a number of trials of medical and even digital merchandise. Or for that matter even unemployment surveys, that are sample-based after which blown as much as clarify the universe. As evidently the samples can’t be greater than half per cent of the inhabitants, we have to ask the query “what if somebody is ignored from this train”. This, in line with the creator, is essential as typically it’s the whole feminine inhabitants that may very well be ignored from the survey or experiment, which is generally the case with any such trials. The identical can maintain for youngsters or the differently-abled and responses are usually skewed. Whereas we want not reject the findings, they is probably not taken at face worth.

One other rule that Harford places throughout is: misinformation may be stunning. This we are able to see after we have a look at firm or authorities shows when charts are proven in a selective method and worded rigorously to indicate info with out disclosing the reality. Utilizing the correct scales and footage and specializing in a couple of achievements, the message conveyed may be very totally different from the true state of the nation or firm.

Even computer systems and algorithms may be deceptive and there have been instances within the USA the place they’ve supposedly scored over scientific analysis when it comes to predicting the unfold of influenza. By utilizing Google to trace the variety of hits which seek for pharmacies or info on the flu, programmes have been capable of predict the unfold of illness in particular areas. Fascinating as it could sound, these algos have limitation within the sense of not with the ability to learn the thoughts by such monitoring. Therefore there have additionally been cases of false reporting, as a search could also be out of curiosity moderately than being bothered by the illness, as has been the case with the Covid pandemic.

At a private degree he warns us to maintain an open thoughts and never be dogmatic on our views on information actions. Right here he provides the instance of the genius economist, Irving Fisher, who made some huge cash predicting future occasions however refused to vary his view because the Melancholy set in and misplaced his whole wealth and have become a giant debtor! This was the results of being dogmatic in views.

It’s true that we reside in a world the place there’s a plethora of information that’s being thrown at us to indicate that issues are actually nice. The language and information level to indicate that the Indian financial system is recovering from the pandemic is just not very totally different from what we noticed within the USA earlier than the presidential elections. That is the place Harford’s 10 guidelines can assist us sieve the cloud of noise and see the true image.

That’s why he ends the ebook by saying that we’ve to continuously stay curious as a result of typically “we predict we all know which we don’t” and this frailty is utilized by the purveyor of data to drive dwelling the messages that is probably not proper.

Madan Sabnavis is chief economist, CARE Ratings

The way to Make the World Add Up: Ten Guidelines for Pondering In another way About Numbers
Tim Harford
Hachette
Pp 338, Rs 699